legality/practicality of nesting complete, valid HTML pages

[henry]

Folks:

Thank you all for your replies. I'll reply briefly to each key point:

My gut feeling is that if you can get past the feeling that this
solution is "fragile" you will be fine. Given that you want to work in
DW, it is the solution that is available to you. Use it, and don't be
afraid of it being fragile.

Thanks! You are probably correct. I wanted to be aware of other options,
that's all. Part of what got me in this mess is not finding out enough
before I jumped in.

[about the frequency of using includes]
It is very common indeed. The inclusion can be done by several
different mechanisms and it is often done off-line -- i.e. before the
page it uploaded using a template mechanism or macro processor.

Thanks, I checked out the template mechanism and it seemed cumbersome.
Using a generalized macro processor seems as if it might put me on the
command line more than I'd like to be. The PHP include mechanism seems just
right.

Well, I like my pages to validate, but it can be overrated from a
practical point of view. If you have existing content that works
exactly as you'd like I would probably have other things on my to do
list before making old pages validate.

Thank you for sharing your experience with validation.

In retrospect, clearly, I should have started by reducing all the existing
content to plain text and rebuilt _everything_ from scratch. I've ended up
doing that with a lot of intermediate wrong turns, but now the site can be
considered all new and so it seems reasonable to validate.

Here's another issue: As a result of having far too much on my plate, I
don't see how I can test the new site as I should, on multiple browsers and
OS platforms. Judging from the lack of response from the alpha group to home
I referred a reasonably good prototype, I have very little chance of getting
anything at all from a beta group. You may consider this a naïve theory,
but I have no clear alternative: I'm hoping that taking care to produce
fully validated code will help me avoid at least some such problems.

The high-level/low-level distinction is not a very useful one unless
you provide a definition of terms to haggle over. I use the tools I
use (rsync, sitecopy, wget, emacs, Perl, m4) because they make things
easier for me, not because they are some sort of puritan hair shirt
that make me admirable (but then I am not a professional web developer
so maybe I don't count). I have looked over the shoulders of people
using Dreamweaver and I wonder how they put up with it, but I image
they do exactly the same when they see me working. The oddity is that
your chosen tool is not, currently, making things easy for you.

HL versus LL is an interesting discussion on its own, but this site is
something like a year late and I need to concentrate on getting it on the
air.

I use tools such as wget, awk and Jedit when I have need for them. My
apparent difficulties with DW may be largely due to the fact that I started
this project knowing nothing about DW and little about web-authoring. Now I
know more, but as this discussion has demonstrated, the skills I've
developed are lopsided.

That said, at this point, I don't see DW as significantly better or worse
than the dozens of software Integrated Development Environments (IDEs) I've
encountered: adequate at many tasks, very convenient for some work, and
frustrating on other points.

For small sites with content that changes only every now and then I
use a text template mechanism to do the "includes" before uploading
the finished HTML: edit the content, run a build command, run
sitecopy.

Right, that makes sense. I'm hoping to collect significantly more content
in coming years, so I think it is reasonable to go about it differently. I
_think_ my scheme will be good for that.

The trend for larger sites seems to be to move the content to an
online content management system. This main advantage is to put the
content back into the hands of the people who generate and maintain
it.

Like Joomla? I've struggled with Joomla some. So far I find it too
generalized for my taste. I want to have full control, and I don't see how
I could do that with Joomla. But that is yet another independent
discussion.

Some documents start with a XML prelude. Also, can you be sure that
all yours have a DOCTYPE at the top? If not you'd need to be a bit
more flexible.

Thanks for mentioning that. Fortunately, I have complete control over the
chunks, even if I went out into the weeds by building them first with HTML
headers and trailers.

Not sure what you mean by that. PHP makes this sort of thing quite
easy. If you have a programming background, all you'd need is a PHP
reference manual. If not, you'd need to find a tame programmer.

Briefly: What I meant is making sure that there's room to process that one
page that I didn't notice grew to 15 times the size of the rest of the
pages. From what I've seen so far PHP has all sorts of cool, practical
tools and it will probably handle that just fine, possibly totally
auto-magically.

I think you missed the point. The tool is unimportant. But no tool is a
substitution for knowing what you are doing. Unfortunately some ( or
many) WYSIWYG editors either promote that "they can be a substitution
for knowing web design" or user mistakenly assume they are. You can make
valid code with almost any tool (well not really...MS Publisher is one
that comes to mind) but with many WYSIWYG editors it takes effort...and
maybe some hand tweaking.

Can't argue with that. Most of my index.php is handcrafted in DW Code
View, not much different than I'd do with a text editor. Of course my
design may be crappy, but so far it does what I expect of it and it hasn't
really broken.

[Name for this scheme, and are there good alternatives]
Ah... a "template" with "server-side includes". If using PHP then PHP
includes. If SSI then SSI includes, etc.

[recommended search terms] includes template web design tutorial

Thanks! That is just what I wanted to know. I spent some time searching
with terms like those and what I got was mostly sites offering me a
selection of existing templates, free or commercial. Now I know that I was
on the right track...

If you use a WYSIWYG editor to make your "parts" for the include files
then you may have to cut and paste the relevant bits from WYSIWYG
development page (code view) and paste into Notepad or something to make
your includes.

Tasks like this are trivial to do in DW by keeping both the WYSIWYG (Design
View) window and the Code View window open at the same time. When I first
saw that capability I thought, "Uh oh, this isn't going to work at all."
After some experience I'll say it doesn't work perfectly, but it isn't too
bad, either.

If you are using includes, your most unimportant development tool will
be a development web server. You can not do server-side includes with
"file://" protocol.

No problem. I created a dedicated subdomain on the server that hosts the
existing site. I've also had good results from using the built-in Apache
server on MacOS. Can work directly in the server tree -- don't even need to
upload.

[about parsing out header and trailer content at load time]

No, it may have some other info was well. Again RegExp is the tool. BUT
if you have access to all these pages then editing out the parts ahead
of time would be the BEST solution. Parsing is only done when you don't
have access to the source.

Yeah, it is clear I need to wipe the headers and trailers on the host. I'm
not a RegExp expert, but I think I can clean most of the files
semi-automatically. DW supports site-wide searches with RegExp.

The upshot is that the inserts must not have <headerstuff, CSS links,
or even <BODYtags... it must be just the inserted code. You do not
need to validate each insert, only validate the generated code.

Yep, my initial post said something like that.

>
I had a similar issue lately when I was working on a Wordpress site. I
don't (yet) use DW, but I read somewhere recently that there's a setting
in DW which allows you to categorise files as inserts, not full docs.

I haven't seen that. Maybe you are thinking about DW's "design-time style
sheets" mechanism. You tag specific header CSS links as design-time only and
those links get auto-filtered out before files are uploaded to the server.
As I commented in an earlier post, the implementation seems a bit fragile to
me. (Because it is applied manually, on a file-by-file basis: I will
forget!) There's an extension available from www.communitymx.com that
apparently applies this on a per-directory basis. I haven't tried it yet.

----

Thanks to everyone! My questions have been answered and I've learned a lot.

Henry

remove 'zzz'